Evaluation of the LDBio Aspergillus ICT lateral flow assay for … · 2020. 10. 15. · with ABPA across the world, assuming about 2.5% of adults with asthma are affected [3]; ABPA
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
RESEARCH ARTICLE
Evaluation of the LDBio Aspergillus ICT lateral
flow assay for serodiagnosis of allergic
bronchopulmonary aspergillosis
Elizabeth Stucky Hunter1, Iain D. Page1,2, Malcolm D. Richardson1,3, David
W. DenningID1,4*
1 Division of Infection, Immunity and Respiratory Medicine, Faculty of Biology, Medicine and Health,
Manchester Academic Health Science Centre, University of Manchester, Manchester, United Kingdom,
2 North Manchester General Hospital, Pennine Acute Hospitals NHS Trust, Manchester, United Kingdom,
3 Mycology Reference Centre Manchester, Manchester University NHS Foundation Trust, Manchester,
United Kingdom, 4 National Aspergillosis Centre, Manchester University NHS Foundation Trust, Manchester,
solution (provided with kit). The test was read at 30 minutes and results were interpreted both
visually (by ‘eye’) and digitally using the Qiagen ESEQuant LR3 (Lake Constance, Germany)
lateral flow reader. Both reads were conducted by the same user, with the visual reading con-
ducted first to eliminate bias resulting from the digital reading. For visual reads, the test was
determined positive by the appearance of two lines: a blue positive control (“C”) line, and a
black positive test (“T”) line. The appearance of any black line at the “T” marker was consid-
ered positive, as recommended in the manufacturer’s guidelines. Using the LR3 lateral flow
reader, a positive test was defined by detection of peaks (any height) between 46.0–48.0mm
and between 53.5–55.5mm for control and test lines, respectively.
Routine diagnostics
In many cases, Aspergillus-specific IgG and IgE, and total IgE levels were measured on ABPA
patient samples as part of routine clinical care. Testing was carried out by the Manchester Uni-
versity NHS Foundation Trust, Department of Immunology, using the automated Immuno-
CAP Phadia 1000 System.
Statistical analyses
To assess diagnostic performance, we calculated Youden’s J statistic (sensitivity + specificity–
1), likelihood ratios and the diagnostic odds ratio (DOR) [17]. Binomial confidence intervals
(95%) were calculated for sensitivity, specificity, and DOR. Pearson’s chi-square statistic was
used for comparisons between subgroups. Spearman’s rank correlation coefficient (ρ) was
used for correlation between ImmunoCAP titers and ICT results. For all results, a two-tailed
P-value< 0.05 was considered statistically significant.
Results
Patients and sera
Patient characteristics for 106 ABPA patients and 141 diseased controls are shown in Table 1.
Within the ABPA patient group, two subgroups were defined as ‘proven’ or ‘presumed’
according to total serum IgE level (‘proven’: >1000 IU/ml, ‘presumed’: 113–999 IU/ml). Dis-
eased controls consisted of patients diagnosed with asthma and/or bronchiectasis. Subgroups
for analysis were defined as (1) ABPA-S: serological ABPA (no bronchiectasis evident, ±asthma), and (2) ABPA-B: ABPA or control with bronchiectasis ± any other conditions
(including asthma and COPD). Additionally, as part of routine testing, 64 of the 106 ABPA
patients in this study had microbiological culture performed on respiratory samples (sputum),
usually multiple specimens. High volume fungal culture technique was usually used, which has
a higher yield than routine culture [18]. Seven (7) cases yielded no growth. Aspergillus fumiga-tus was the main pathogen isolated in the remaining 57 cases, and 21 of these also grew other
Aspergillus species in culture. The following data was not available for ‘asthma only’ control
subjects: (1) gender, (2) whether sputum culture was performed (and results).
ICT results
For all 106 ABPA serum samples, 96 tested positive by ICT with 90.6% sensitivity (95% CI,
83.3% to 95.4%). Sensitivity for ‘proven’ and ‘presumed’ ABPA subgroups was 89.8% (95% CI,
81.5% to 95.2%) and 94.4% (95% CI, 72.7% to 99.9%), respectively, with no significant differ-
ence between these groups. No associations with any of multiple parameters were linked with
false negative assay ICT assay. In the total diseased control group, 143 of the 164 sera tested
negative by ICT with 87.2% specificity (95% CI, 81.1% to 91.9%) (Table 2).
PLOS ONE LDBio Aspergillus ICT lateral flow assay for serodiagnosis of ABPA
PLOS ONE | https://doi.org/10.1371/journal.pone.0238855 September 25, 2020 4 / 13
Aspergillus fumigatus growth in sputum culture, n (%) 57 (53.8) 48 (54.5) 9 (50.0) 3 (1.9) 0 (0) 3 (4.2)
A. fumigatus only, n (%) 36 (34.0) 28 (31.8) 8 (44.4) n/a 0 (0) n/a
A. fumigatus + other Aspergillus spp., n (%) 21e (19.8) 20 (22.7) 1 (5.6) n/a 0 (0) n/a
Other Aspergillus spp. (only) growth in sputum culture,
n (%)
0 (0) 0 (0) 0 (0) n/a 0 (0) n/a
aAll ABPA: proven + presumedbAll control: asthma ± bronchiectasiscn/a = data not availabledCOPD = chronic obstructive pulmonary disease.eA. niger [17], A. nidulans [3], A. fischeri [1], A. montevidensis [1], A. terreus [1], A. versicolor [1], A. pallidoflavus [1].
https://doi.org/10.1371/journal.pone.0238855.t001
Table 2. Summary of results for LDBio Aspergillus ICT IgG-IgM test and routine serological assay.
Sensitivity Specificity Youden’s DOR
Analysis group % (95% CI) % 95% CI (%) Index (95% CI)
Fig 1. (a) LDBio Aspergillus ICT assay format and (b) Comparison of LDBio Aspergillus ICT result with ImmunoCAP Asp IgG and IgE titers (n = 52). Positive
result cut-off for each assay denoted by dashed line (ImmunoCAP Aspergillus IgG = 40mg/L, ImmunoCAP Aspergillus IgE = 0.35 kUA/L).
https://doi.org/10.1371/journal.pone.0238855.g001
PLOS ONE LDBio Aspergillus ICT lateral flow assay for serodiagnosis of ABPA
PLOS ONE | https://doi.org/10.1371/journal.pone.0238855 September 25, 2020 7 / 13
found 20% prevalence among Brazilian asthmatic patients [19], and several studies in India
found ABPA in approximately 5% [20] to 8% [5, 21], and up to 20% [22] of asthmatics, with
up to half of patients being misdiagnosed as pulmonary tuberculosis [9]. The incidence of
ABPA is also higher in patients with severe asthma and/or corticosteroid-dependent asthma
(7–14%), and in patients with atopy [23]. Diagnosis of ABPA relies on a body of clinical, radio-
logical, and immunological (and/or mycological) evidence; and the degree of difficulty in diag-
nosing ABPA largely depends on the staging and severity of disease. Diagnosis is relatively
straightforward when a patient exhibits all clinical symptoms including bronchiectasis, and
most patients are diagnosed at this stage [24]. Earlier diagnosis, however, is ideal to prevent
development of bronchiectasis and irreversible tissue damage [25]. Serological tests are a useful
tool for early detection of ABPA and comprise a significant portion of the recently suggested
guidelines for ABPA diagnosis [10], with raised total IgE and Asp IgE (or positive skin prick
test) being ‘major’ criteria and raised Asp IgG or positive precipitins test included under
‘minor’ criteria.
The LDBio Aspergillus ICT lateral flow assay is a new diagnostic test requiring minimal
time and resources. Our recent evaluation in CPA patients (vs. healthy controls) determined
the test to have good sensitivity (91.6%) and specificity (98.0%), and a significant improvement
in performance over our current workhorse assay (ImmunoCAP EIA) to detect raised Asp IgG
[16]. The current evaluation in ABPA patients and diseased controls from the United King-
dom has shown the Aspergillus ICT to have good overall sensitivity (90.6%) and specificity
(87.2%) in distinguishing patients with ABPA from those with underlying respiratory disease.
Not surprisingly, the test exhibited significantly better sensitivity for ABPA with bronchiectasis
(ABPA-B) (93.9%) than for serological ABPA (ABPA-S) ± asthma (no evidence of bronchiec-
tasis) (79.2%). Bronchiectasis is associated with progression of ABPA and a more severe form
of disease [1]. Persistent Aspergillus infection may lead directly to bronchiectasis or drive
recurrent exacerbations and progression of bronchiectasis already present due to ABPA or
severe asthma [26, 27]. Once present, the pathophysiology of bronchiectasis can facilitate per-
sistent infection through impairment in mucociliary clearance [28] and structural/tissue dam-
age that creates a permissive environment for establishment of infection [26, 29, 30]. Patients
with bronchiectasis (with or without ABPA) may develop Aspergillus bronchitis, and this
entity is also associated with raised Asp IgG antibodies [31]. We did not explicitly assess a
group of asthmatic patients with Aspergillus sensitisation for Aspergillus IgG with the ICT or
ImmunoCap, and given the likely interaction with Aspergillus bronchitis, we would need to do
so knowing if Aspergillus bronchitis was or was not present.
All cases in this study with Aspergillus species growth in high-volume sputum culture were
positive for Aspergillus fumigatus and approximately one-third (37%) were also positive for
other Aspergillus species. There was no significant difference in ICT performance between
cases with A. fumigatus growth alone versus those with other species present. There were no
cases in this study with non-A. fumigatus ABPA, however, a limited number of non-A. fumiga-tus CPA cases were evaluated in a previous study and we found no significant difference in
ICT performance between A. fumigatus and non-A. fumigatus CPA cases [16]. The ability to
detect antibody to non-A. fumigatus may be of particular importance in regions where ABPA
due to non-A. fumigatus species is more prevalent [32]. For the diagnosis of ABPA, sputum
culture is considered a supportive (not diagnostic) test [10]. In Indian patients thought to have
ABPA, sputum tested by microscopy and culture was positive for Aspergillus species in only
32.5% of 203 cases [8]. Higher volume cultures yielded a positive in 56% of 75 cases (similar to
this series), some of whom were on antifungal therapy compared with conventional culture
(12.5%) [18]. Aspergillus ICT was more sensitive than this. Culture has the advantage of being
PLOS ONE LDBio Aspergillus ICT lateral flow assay for serodiagnosis of ABPA
PLOS ONE | https://doi.org/10.1371/journal.pone.0238855 September 25, 2020 8 / 13
able to check antifungal susceptibility, if positive, although direct detection of resistance is now
possible [33–36].
Given the ‘major’ and ‘minor’ requirements of both raised Asp IgE and IgG for ABPA diag-
nosis, we sought to determine if the homogeneous antigen-sandwich format of the AspergillusICT rendered it capable of detecting IgE (in addition to IgG and IgM as claimed) and whether
the test displayed an isotype bias. Although we found a significant correlation between levels
of Asp IgG and Asp IgE, we did not observe any such isotype bias using the ICT. As expected
based on our previous findings [16], we found the test to perform well at increasing levels of
Asp IgG (ImmunoCAP ‘positive’, >40mgA/L) regardless of Asp IgE titer. The test also per-
formed well at raised levels of Asp IgE paired with values of Asp IgG considered ‘negative’ by
the recommended ImmunoCAP cut-off (� 40mgA/L). While this may possibly indicate the
Fig 2. Interpretation of Aspergillus serology and clinical presentation of ABPA for diagnosing Aspergillus disease: allergic
bronchopulmonary aspergillosis (ABPA), Aspergillus rhinosinusitis (RS), severe asthma with fungal sensitization (SAFS),
detection of Asp IgE by the ICT test format, we have previously found this assay to detect Aspantibody in patients with clinically confirmed CPA who have tested ‘negative’ by ImmunoCAP
Asp IgG [16, 37]. Further studies would be necessary to evaluate the individual contributions
of IgG and IgE from patient sera and the antibody class-specific performance of the AspergillusICT.
Individual serological tests alone for total IgE or Asp IgE or IgG are not specific for ABPA.
It is important to note that routine tests used for ABPA diagnosis—raised Asp IgE, IgG and/or
positive precipitins test—are also used in the diagnostic pathways for Aspergillus diseases other
than ABPA including Aspergillus bronchitis (AB), acute and sub-acute invasive aspergillosis
(IA), CPA, severe asthma with fungal sensitization (SAFS), and chronic or granulomatous
Aspergillus rhinosinusitis (RS) [12]. The diagnostic interpretation of these tests, alone or in
combination with each other and/or clinical presentation is summarized in Fig 2. Under gen-
erally accepted diagnostic criteria for Aspergillus disease, ABPA is the only condition using the
combined requirements of raised Asp IgG and IgE. In practice however, it is not uncommon
to see raised total and/or Asp IgE in CPA patients [38], especially those that have developed
CPA as a result of untreated ABPA progression [39]. Additionally, although a raised level of
Asp IgG is considered an exclusion criteria for SAFS, approximately 10% of all asthmatics pres-
ent with raised Asp IgG [1] and there are likely to be cases of SAFS with increased levels of
both Asp IgG and IgE, but lacking the clinical and radiological features of ABPA [12]. High
titers of total serum IgE are encountered not only in ABPA, but also in cases of SAFS, Aspergil-lus rhinosinusitis, and in atopic asthmatics [40, 41]. Furthermore, cut-off values to distinguish
these conditions are speculative and may even be different in ABPA complicating asthma ver-
sus ABPA complicating CF [4, 42].
In this study of clinically confirmed cases of ABPA compared to diseased controls, we
found the LDBio Aspergillus ICT to have good sensitivity and specificity. The test effectively
distinguished between Aspergillus-sensitization complicating asthma and/or bronchiectasis,
and underlying conditions. It is rapid (result in<30 minutes) and easy to perform, with simple
result interpretation by visible inspection. Overall, the LDBio Aspergillus ICT exhibits excellent
performance as a screening tool in the ABPA diagnostic pathway.
Supporting information
S1 Dataset.
(XLSX)
Acknowledgments
Diseased control sera were provided by the ManARTS research tissue bank, Wythenshawe
Hospital, Manchester University NHS Foundation Trust (Manchester, UK).
Author Contributions
Conceptualization: Malcolm D. Richardson, David W. Denning.
Data curation: Iain D. Page.
Formal analysis: Elizabeth Stucky Hunter.
Funding acquisition: David W. Denning.
Investigation: Elizabeth Stucky Hunter, David W. Denning.
Methodology: Elizabeth Stucky Hunter.
PLOS ONE LDBio Aspergillus ICT lateral flow assay for serodiagnosis of ABPA
PLOS ONE | https://doi.org/10.1371/journal.pone.0238855 September 25, 2020 10 / 13